Renin-Angiotensin-Aldosterone System, RR Interval ...

AJH 1995; 8:683-688

ORIGINAL CONTRIBUTIONS

Renin-Angiotensin-Aldosterone System, RR Interval, and Blood Pressure Variability During Postural Changes in Borderline Arterial Hypertension Daniel A. Duprez, Johan H. De Sutter, Marc L. De Buyzere, Ernst R. Rietzschel, Steven Rimbaut, Jean M. Kaufman, Marie-Jos~ Van Hoecke, and Denis L. Clement

This study aimed to examine the changes in RR interval and blood pressure (BP) variability and humoral factors during postural changes in borderline arterial hypertension. Twenty-nine patients (44 ± 3 year; BP, 145 3/84 _ 3 mm Hg) with borderline hypertension and a control group of 38 subjects (35 - 2 year; BP, 123 ± 3/70 ± 2 mm Hg) underwent power spectrum analysis of RR interval and BP (low frequency (LF), 0.05 to 0.15 Hz; high frequency (HF), 0.15 to 0.40 Hz) in the supine and standing positions. Concentrations of plasma renin activity, angiotensin II, and aldosterone in supine and standing positions had been determined in the hypertensive group. Borderline hypertensives are characterized by higher oscillations of systolic and diastolic BP, but not of RR interval in the supine position versus the control group. Low frequency and HF components of systolic and diastolic BP expressed as absolute data are also significantly higher in borderline hypertensives. Moreover, standing tended

to increase the LF/HF ratio of both RR interval and BP variability compared to controls. The standing position was able to further activate the LF but not the HF component of BP variability. In borderline hypertension renin release during postural changes correlated well with the decrease in the power of the HF vagal component of RR interval variability (r = -0.70, P < .001) and with the increase of the LF component of diastolic blood pressure variability (r = 0.43, P = .03). In conclusion, our results indicate that in borderline arterial hypertension, LF and HF oscillations of BP are already significantly increased at rest. Standing was capable of significantly activating the LF component, but not the HF component, of BP variability. Am J Hypertens 1995;8:683-688

t has been recognized for several decades that blood pressure (BP) has the tendency to fluctuate with significant amplitudes around a central set point. Mayer observed that these fluctuations

were rhythmic in nature and oscillate at frequencies slower than the respiratory rhythm.1 Furthermore, it has been suggested that these oscillations reflect sympathetic vasomotor tone and rhythmic changes of peripheral vascular resistance. 2 During the past decade calculation of the power spectrum of data collected from continuous registration of BP and RR interval allowed analysis of these rhythmic circulatory p h e n o m e n a . 3'4 Two major rhythmic oscillations are recognized in the power spectrum of both heart rate (RR interval) and BP vari-

I

Received February 24, 1994. Accepted January 31, 1995. From the Departments of Cardiology, Angiology, and Endocrinology, University Hospital, Gent, Belgium. Address correspondence and reprint requests to Daniel Duprez, MD, DSc, PhD, Department of Cardiology and Angiology, University Hospital, De Pintelaan 185, B-9000 Gent, Belgium.

KEY WORDS: Borderline arterial hypertension, plasma renin activity, aldosterone, postural changes, power spectral analysis.

Downloaded from https://academic.oup.com/ajh/article-abstract/8/7/683/216311/Renin-Angiotensin-Aldosterone-System-RR-Interval © 1995 by the American Journal of Hypertension, Ltd. by University of Minnesota Libraries - Twin Cities user on 12 September 2017

0895-7061/95/$9.50 0895-7061(95)00080-9

684

DUPREZET AL

AJH-JULY 1995-VOL. 8, NO. 7

ability in the healthy subject at rest. 3"5 The low frequency (LF) components cover the frequency interval of 0.04 to 0.15 Hz, and the high frequency (HF) components are considered in the interval between 0.15 and 0.40 Hz. At present there is some evidence that the HF spectral component of RR interval variability indicates vagal modulation and its LF component is a marker of sympathetic modulation. The group of Malliani and Lombardi described that the LF/HF ratio is a characteristic of sympathicovagal interactions. 4'6 They found that all maneuvers enhancing sympathetic drive (eg, 90 ° upright tilt) increased LF and decreased HF of the RR interval variability, and as a consequence there is a highly significant increase in the LF/HF ratio. Concerning BP spectral indices, LF power of systolic variability has been advocated as a marker of sympathetic vasomotor tone. 4 The early stage of hypertension is characterized by increased sympathetic activity and heart rate. 7's In established hypertensives compared to age-matched normotensives higher LF and lower HF components of RR interval variability have been observed indicating an altered sympathetic vagal balance. 6 In essential hypertension LF and HF components of BP variability and effects of postural changes are less well recognized. Another hallmark of essential hypertension is the alteration of the renin-angiotensin-aldosterone (RAA) axis. Interactions between the RAA axis and the orthosympathetic nervous system are know. The RAA axis induces a-adrenergic vasoconstriction, whereas ~-adrenergic activation further stimulates renin secretion. 9-1~ The question may arise whether the RAA system plays an important role in humans in shortterm cardiovascular regulation as manifested by its ability to modulate the amplitude of spontaneous fluctuation in heart rate and BP.12 Thus, the present study aimed: 1) to document hormonal changes in RAA axis and spectral indices of RR interval and BP variabilities brought about by postural maneuvers and 2) to investigate if the release of humoral factors of the RAA axis closely reflects the changes of spectral indices of RR interval and BP variability induced b y p o s t u r a l m a n e u v e r s in b o r d e r l i n e arterial hypertension. METHODS Patients Twenty-nine patients (13 men, 16 women) with a mean age of 44 + 2.5 years (mean -+ SEM), a mean height of 1.72 + 0.02 m, and a mean weight of 74.8 -+ 2.7 kg with borderline essential hypertension were enrolled in this study protocol. These patients were determined to have borderline arterial hypertension on the basis of mean office arterial BP readings measured on three separate examinations during 6 months. History, physical examination, or labora-

tory data did not provide evidence of secondary forms of hypertensive disease in any patient. Patients had never taken any drug treatment. Concomitant cardiovascular diseases and other chronic diseases, such as diabetes mellitus, liver disease, and renal disease, were exclusion criteria as well. All patients were nonsmokers and refrained from drinking coffee and alcohol at least 24 h before the examination. All patients gave oral informed consent to participate in this study. The study protocol had been approved by the local medical ethical committee. Controls Thirty-eight apparently healthy subjects, who did not take any drugs (mean age, 35 + 2 years; mean height, 1.68 - 0.02 m; mean weight, 65.4 + 2.0 kg) were considered as a control group for the analysis of RR interval and BP variability by power spectrum analysis in the supine and standing positions. None of them was a smoker and they all refrained from drinking coffee and alcohol at least 24 h before the examination. Arterial BP and RR interval Variability In each subject beat-to-beat BP was measured by means of a Finapres 2300 (Ohmeda, Englewood, CO) device working on the volume clamp method described by Penaz.13 The Finapres operates through a cuff wrapping the middle phalanx of the third finger, and BP is recorded by a photoplethysmograph measuring arterial blood flow. This device provides continuous BP values that correlate with intraarterially recorded BP values under a variety of testing conditions. 14 The BP was first monitored for 30 min in the supine position. D u r i n g BP r e c o r d i n g s , a s e r v o selfadjustment operates, periodically testing the photoplethysmograph's set point. Blood pressure recording was then started, after switching off the servo self-adjustment system. Supine BP recording was prolonged to obtain at least two stationary sequencies of 256 beats. An additional sequence was recorded during active standing after 30 min. The RR interval was derived from the beat-to-beat recordings from the electrocardiogram in the same manner. Data Analysis Data of RR interval and systolic and diastolic BP were fed into a CARDIO-PSA software package (Medatec, Brussels, Belgium) to calculate variances of RR interval and BP and their spectral indices: LF, HF, and LF/HF ratio. Nonstationary segments (including erratic and sudden BP and RR interval variations) were automatically discarded. The spectral characteristics of the remaining segments were estimated with a fast Fourier transform in a sequential fashion, using a sample length of 256 beats. Cumulative powers were computed over two fre-

Downloaded from https://academic.oup.com/ajh/article-abstract/8/7/683/216311/Renin-Angiotensin-Aldosterone-System-RR-Interval by University of Minnesota Libraries - Twin Cities user on 12 September 2017


quency bands, the so-called LF (0.04 to 0.15 Hz) and the HF (0.15 to 0.40 Hz) bands. For each subject LF and HF were calculated in absolute values (msec 2, mm Hg 2) integrating the density curves in the respective frequency ranges and then averaged to obtain mean values for the group as a whole. Moreover, normalized units (nu) are defined as the component power divided by the total oscillatory power. Blood samples were taken at 8:30 Ara on the day of the BP and RR interval measurements; the sample was withdrawn from the brachial vein after the patient fasted overnight and was supine for 30 min. Another blood sample was taken after 30 min of standing. Plasma renin activity was measured using a commercial radioimmunoassay from Pharmacia Diagnostica (Phadebas Angiotensin I test, Uppsala, Sweden). Angiotensin II was estimated by a radioimmunoassay procedure (Amersham International, Amersham, Buckinghamshire, England) after isolation from plasma by solvent extraction chromatography (Analytichem Bond Elut, Harbor City, CA) and purification by HPLC (Nucl 10-C18 column). Aldosterone was assayed with a commercial radioimmunoassay from Radim (Pomezia, Rome, Italy). Hormone Measurements

Statistical A n a l y s i s All study parameters are expressed as the mean value -+ standard error of the mean. The difference between the mean values of the studied parameters were compared with a Wilcoxon test for both controls and patients. Changes of the power spectral analysis of BP and RR interval variability and the humoral factors were intercorrelated by the Spearman rank test. Further relations between the study parameters were evaluated by linear regression analysis, and regression lines were compared by analysis of covariance. Probabilities at P < .05 were considered significant in this study. RESULTS Spectral Indices of BP and Heart Rate Variability D u r i n g Postural C h a n g e s in Controls and Borderline H y p e r t e n s i v e s Heart rate and arterial BP are given

in Table 1. When changing from the supine to the standing position, heart rate increased from 73 + 2 to 82 -+ 2 beats/min (P < .001) in the hypertension group and from 77 + 2 beats/min to 90 - 2 beats/min in the control group. Systolic and diastolic BP did not change significantly. The variances of RR interval and arterial BP in controls and in borderline hypertensives during postural changes are summarized in Table 2. In borderline hypertension, total variance of the RR interval decreased going from supine to standing from 2461 -891 msec 2 to 1681 _+ 442 msec 2 (P < .05). There was a slight increase of the LF component but not significant, whereas the HF component decreases signifi-

R A A AXIS AND POWER SPECTRUM ANALYSIS

685

TABLE 1. HEMODYNAMICS DURING POSTURAL CHANGES IN CONTROLS VERSUS BORDERLINE ARTERIAL HYPERTENSION (MEAN + SEM) Variables C ont rol H e a rt rate (beats/min) Systolic b l o o d p r e s s u r e ( m m Hg) Diastolic b l o o d p r e s s u r e ( m m Hg) Borderline H y p e r t e n s i o n H e a r t rate (beats/rain) Systolic bl ood p r e s s u r e ( m m Hg) Diastolic bl ood p r e s s u r e ( m m Hg)

Supine

Standing

77 --- 2 123 +-- 3 70 ± 2

90 + 2* 119 + 2 66 ± 2

73 ± 2 145 ± 3 84 ± 3

82 ± 2* 139 ± 3 84 ± 2

*P < .01,

cantly from 923 + 329 to 311 + 76 (P < .001). During the change in posture, there was a significant increase (P < .001) of the total variance of systolic BP from 29.3 --- 4.1 mm Hg 2 to 51 + 8.4 m m Hg 2 and the LF component of systolic BP from 13.3 + 1.5 mrn Hg 2 to 32.1 -+ 5.5 mm Hg 2, whereas there was no significant change in the HF component. A similar increase was seen for total variance of diastolic BP, namely from 9.9 + 1.9 mm Hg 2 to 15.9 -+ 2.0 mm Hg 2 (P < .001) and for the LF component, from 4.9 +- 0.6 m m Hg 2 to 10.8 + 1.4 mm Hg 2. There was a nonsignificant increase in the HF component. Total variance of RR interval and its LF and HF component were comparable between borderline hypertensives and controls in the supine and standing positions. On the contrary, total variance of systolic (P < .01) and diastolic (P < .01) BP in supine position were significantly higher in borderline hypertensives versus controls. The same held true for LF and HF components (P < .01). Going from supine to standing, mean increase of total variance of systolic and diastolic BP was comparable between both groups. This increase in variance was reflected in the LF and HF components in the controls, but only in the LF TABLE 2. VARIANCES OF RR-INTERVAL A N D ARTERIAL BLOOD PRESSURE IN BORDERLINE ARTERIAL HYPERTENSION DURING POSTURAL CHANGES

RR interval Total variance, m s e c 2 LF c o m p o n e n t , m s e c 2 HF c o m p o n e n t , m s e c 2 Systolic b l o o d p r e s s u r e Total variance, m m H g 2 LF c o m p o n e n t , m m H g 2 HF c o m p o n e n t , m m H g 2 Diastolic b l o o d p r e s s u r e Total variance, m m H g 2 LF c o m p o n e n t , m m H g 2 HF c o m p o n e n t , m m H g 2

Supine

Standing

2461 + 891 703 + 145 923 + 329

1681 ± 442* 892 + 175 311 + 76~"

29.3 ± 4.1 13.3 - 1.5 7.8 --- 1.5

51.0 + 8 . 4 t 32.1 ± 5 . 5 t 8.5 + 1.2

9.9 ± 1.9 4.9 -+ 0.6 2.1 +- 0.7

15.9 ± 2 . 0 t 10.8 +- 1 . 4 t 2.2 ± 0.3

*P < ,05; fP < .001.


686

DUPREZ ET AL


component for borderline hypertensives. Standing decreased total variance of RR interval significantly more in hypertensives than in controls (P < .05). Table 3 summarizes the spectral indices expressed as normalized units of RR interval and arterial BP variabilities during postural changes in controls and in borderline arterial hypertension. In borderline hypertensives, going from supine to standing the LF of the RR interval increases from 42.6 ± 3.8 nu to 59.4 -+ 3.5 nu (P < .001). The HF decreased from 38.2 ± 3.1 nu to 20.5 ± 2.3 nu (P < .001). There was an increase of the LF/HF ratio as welt, namely from 1.61 - 0.31 to 4.56 ± 0.94 (P < .001). The LF of systolic BP increased from 49.4 ± 2.5 nu to 61.7 + 2.8 nu (P < .001) and also the LF/HF ratio from 2.44 + 0.31 nu to 4.24 ± 0.58 nu (P < .001). There was no significant change of the HF. As for the diastolic BP during the postural changes, there was a significant increase of the LF, namely from 57.7 + 2.8 nu to 69.0 +- 2.4 nu (P < .001), and also the LF/HF ratio increased from 4.36 ± 0.60 nu to 6.36 _+ 0.84 nu (P < .05). There was no change of the HF component. The LF/HF ratio was comparable in supine position for RR interval and systolic and diastolic BP variabilities in controls and hypertensive patients. In the standing position, the LF/HF ratio of RR interval (+183% v +106%), systolic BP (+73% v +28%), and diastolic BP (+ 46% v + 38%) tended to be higher for borderline hypertensives. However, LF and HF components of RR interval and BP variabilities expressed as normalized units were not significantly different between controls and patients, either in the supine or in the standing position. Correlations between RAA Axis and Spectral Indices of RR Interval and BP Variability During Postural Changes in Borderline Hypertension The humoral factors increased during postural changes: plasma reTABLE 3. SPECTRAL INDICES OF RR-INTERVAL AND ARTERIAL PRESSURE VARIABILITIES DURING POSTURAL CHANGES IN BORDERLINE ARTERIAL HYPERTENSION

RR interval LF (nu) HF (nu) LF/HF ratio Systolic blood p r e s s u r e LF (nu) HF (nu) LF/HF ratio Diastolic blood p r e s s u r e LF (nu) HF (nu) LF/HF ratio nu, normalized units. *P < .05; fP < .001.

Supine

Standing

42.6 ± 3.8 38.2 ± 3.1 1.61 ± 0.31

59.4 -+ 3.Si20.5 + 2.3i4.56 -+ 0.94f

49.4 ± 2.5 26.2 -+ 2.2 2.44 -+ 0.31

61.7 + 2.8f 20.4 ± 2.0 4.24 ± 0.58t

57.7 ± 2.8 18.4 ± 1.7 4.36 ± 0.60

69.0 ± 2.4t 14.1 +- 1.1 6.36 ± 0.84*

nin activity from 2.26 _+ 0.45 to 3.59 + 0.55 ng/mL/h (P < .05), angiotensin II from 4.23 + 0.69 to 6.03 0.82 pg/mL (P < .05), and aldosterone from 12.41 --1.36 to 26.74 ± 2.65 ng/dL (P < .001). Increases in PRA during postural maneuvers correlated with the decrease in the power of the HF component of RR interval variability (r = -0.70, P < .001) and with the increase of the LF component of diastolic BP variability (r = 0.43, P = .03). Angiotensin II and aldosterone were not statistically significantly correlated with changes of spectral indices of BP and heart rate variability d u r i n g postural maneuvers. DISCUSSION In this study, evidence has been provided that borderline hypertensives were characterized by higher oscillations of systolic and diastolic BP but not of RR interval in supine position, versus controls. The same held true for absolute measures of LF and HF components of systolic and diastolic BP variabilities. Standing tended to increase the LF/HF ratio of RR interval and BP variability in borderline hypertensives versus controls. Our results further indicate that during postural maneuvers in borderline essential arterial hypertension, there is a clear relation between the RAA axis and the sympathetic and vagal modulations of heart rate and BP oscillations, expressed here as spectral indices of their power spectrum. Particularly, renin release from supine to standing position is correlated with the decrease of vagal modulation of RR interval and with the increase of sympathetic modulation of diastolic blood pressure. Power Spectral Analysis and Postural Changes In our study we demonstrated that going from supine to standing, borderline arterial hypertension is characterized by a decrease of total RR interval variability compared to controls, which is mainly due to a decrease of the HF or vagal component. On the contrary, systolic and diastolic BP variability increase upon standing, which is largely due to the increase in the LF or sympathetic components. These phenomena hold for both analyses based on absolute and normalized values. Standing was able to further activate the LF but not the HF component of BP variability in borderline hypertensives. Guzetti et al ~5 studied heart rate variability in normotensive (n = 40), borderline (n = 16), and established hypertensive subjects (n = 40) at rest and after passive tilting; however, BP variability was not a considered study parameter. Our results were in agreement with these investigators' observations of RR interval variability. In established hypertensive patients RR interval variability at rest shows a greater LF and a smaller HF component compared to normotensives,



suggesting enhanced sympathetic and reduced vagal activity. 16 Furthermore, in established hypertensives response to passive tilting is blunted and a significant correlation with the hypertensive state has been suggested. However, Julius et al 7"8 described that increased heart rate is more evident in patients with borderline arterial hypertension. This increase may be due to a higher sympathetic tone, but only a weak correlation was found between levels of heart rate and plasma catecholamine concentration. 17 On the other hand, a decreased parasympathetic activity in borderline hypertension has been reported. 7 Our results indicate only a moderate reduction of the HF component of RR interval variability in supine and standing position.

Methodologic Considerations

Omboni et a114 demonstrated that noninvasive beat-to-beat BP recordings (Finapres at the level of the finger) represent an acceptable substitute of invasive recordings for the analysis of BP spectral components. However, LF power of systolic BP may be slightly overestimated by analysis of noninvasive finger BP recordings. Also Antivalle et a118 have shown that noninvasive BP recordings from the finger can be used reliably to study short-term BP variability by spectral analysis. Our results for systolic BP variability measured by a noninvasive finger BP recording are in accordance with the findings of Furlan et a119 who measured systolic arterial BP by an intraarterial microtip transducer at rest and during tilt in a group of moderately arterial hypertensive subjects. Total variance of systolic arterial BP and the LF component increased significantly from rest to passive tilt. This LF component (Mayer waves) appears to represent a good marker of the sympathetic activity impinging on arterial smooth muscle tone, whereas the HF of arterial pressure, which is largely influenced by mechanical changes, cannot be considered a marker of vagal activity. 2° Concerning short-term diastolic BP variability Pagani et al 4 studied seven normotensive subjects during passive tilt in w h o m they found a significant increase in total power and LF component of diastolic BP and a significant decrease of HF component. In our group of borderline hypertensive subjects there was no significant decrease of the HF component of the diastolic BP. RAA Axis and Neural Stimuli It is well known that plasma renin activity may differ in relation to changes in posture. Brown et a121 observed an increase in plasma renin concentrations of 64% induced by tilting in seven normal subjects. These increases were too high to be explained by volume changes only. Bunag et al9 suggested that renin can be released by neural stimuli. The sympathetic and parasympathetic nervous systems are usually considered to be the principal sys-

RAA AXIS A N D POWER SPECTRUM ANALYSIS

687

terns involved in short-term cardiovascular control on the time scale of seconds to minutes. It has been suggested that renin-angiotensin system could also play a role in short-term cardiovascular control. Akselrod et al 3'22 found in an animal model that there is a possibility that the tonic activity of the renin-angiotensin system normally damps the amplitude of these fluctuations in peripheral vasomotor tone. Our study now indicates a highly significant inverse correlation between the increase in plasma renin activity and the decrease in the vagal component of the RR interval variability during postural maneuvers. Moreover the renin release during postural changes is positively related to the increase of the sympathetic component of diastolic BP variability. Therefore, in borderline arterial hypertension, postural maneuvers induce renin release associated with vagal modulation of the heart rate oscillations and with sympathetic modulation of the diastolic BP oscillations. This gives evidence that the RAA system is also involved in the short-term control of the RR interval and BP variability in borderline arterial hypertension. Our findings should be interpreted in view of the synergism between the RAA system and the sympathetic nervous system in maintaining BP during standing. To demonstrate a causal role of the RAA system in the spectral changes during postural maneuvers future studies with angiotensin-converting enzyme inhibitors or angiotensin II receptor antagonists to abolish the effect should be considered. ACKNOWLEDGMENTS We thank Mrs. Brusselmans and Mrs. Maas for their technical assistance and Mrs. Packet for typing the manuscript. REFERENCES 1. Penaz J: Mayer waves: history and methodology. Automedica 1978;2:135-141. 2. Appel ML, Berger RD, Saul JP, et al: Beat to beat variability in cardiovascular variables: noise or music? J Am Coll Cardiol 1989;14:1139-1148. 3. Akselrod S, Gordon D, Ubel F, et al: Power spectrum analysis of heart rate fluctuations: a quantitative probe of beat-to-beat cardiovascular control. Science 1981; 213:220-223. 4. Pagani M, Lombardi F, Guzetti S, et al: Power spectrum analysis of heart rate and arterial pressure variabilities as a marker of syrnpathico-vagal interaction in man and conscious dog. Circ Res 1986;59:178-193. 5. Koizumi K, Terui N, Kollai M: Effect of cardiac vagal and sympathetic nerve activity on heart rate in rhythmic fluctuations. 1 Aut Nerv Syst 1985;12:251-259. 6. Guzzetti S, Piccaluga E, Casati R, et al: Sympathetic predominance in essential hypertension. A study employing spectral analysis of heart rate variability. J Hypertension 1988;6:711-717. 7. Julius S, Pascual AV, London R: Role of parasympathetic inhibition in the hyperkinetic type of borderline hypertension. Circulation 1971;44:413-418.


688

DUPREZET AL

Julius S, Esler M: Autonomic nervous cardiovascular regulation in borderline hypertension. Am J Cardiol 1975;36:685-696. 9. Bunag RD, Page IH, McCubbin JW: Neural stimulation of release of renin. Circ Res 1966;19:851-858. 10. Gordon R, Kiichel O, Liddle G, Island D: Role of the sympathetic nervous system in regulating renin and aldosterone production in man. J Clin Invest 1967;46: 599-605.


8.

11.

Williams G, Rose L, Dluhy R, et al: Abnormal responsiveness of the renin aldosterone system to acute stimulation in patients with essential hypertension. Ann Int Med 1970;72:317-326.

12.

Pomeranz B, Macauley JG, Caudill MA, et al: Assessment of autonomic function in humans by heart rate spectral analysis. Am J Physiol 1985;248:H151-H153.

13.

Molhoek GP, Wesseling KH, Settels JJM, et al: Evaluation of the Penaz servo-plethysmo-manometer for the continuous, non invasive measurement of finger blood pressure. Basic Res Cardiol 1984;79:598-609. Omboni S, Parati G, Frattola A, et al: Spectral and sequence analysis of finger blood pressure variability. Comparison with analysis of intra-arterial recordings. Hypertension 1993;22:26--33.

14.

15.

16.

Guzetti S, Piccaluga E, Casati R, et al: Sympathetic predominance in essential hypertension: a study employing spectral analysis of heart rate variability. J Hypertens 1988;6:711-717. Malliani A, Pagani M, Lombardi F, Cerutti S: Cardio-

17.

18.

19.

20.

21.

22.

vascular neural regulation explored in the frequency domain. Circulation 1991;84:482-492. Agabatti-Rosei E, Alicandri C, Fariello R, et al: Plasma catecholamines in essential hypertension. J Clin Pharmacol Res 1982;II(suppl 1):69-74. Antivalle M, Salvaggio A, Lattuada A, Libretti A: Analysis of short-term blood pressure fluctuations recorded non-invasively, in Di Rienzo M, Mancia G, Parati G, Pedotti A, Zanchetti A (eds): Blood Pressure and Heart Rate Variability. Amsterdam, IOS Press, 1992, pp 102-108. Furlan R, Dell'Orto S, Cuvellaro W: Effect of tilt and treadmill exercise on short-term variability in systolic arterial pressure in hypertensive men. J Hypertens 1987;5(suppl 5):$423-$425. Malliani A, Pagani M, Lombardi F, et ah Spectral analysis to assess increased sympathetic tone in arterial hypertension. Hypertension 1991;17(suppl III):III36III42. Brown JJ, Davies DL, Lever AF, et ah Plasma renin concentration in relation to changes in posture. Clin Sci 1966;30:279-284. Akselrod S, Oz O, Eliash S: Neural and humoral factors in regulation of blood pressure in the investigation of essential hypertension, in Di Rienzo M, Mancia G, Parati G, Pedotti A, Zanchetti A (eds): Blood Pressure and Heart Rate Variability. Amsterdam, IOS Press, 1992, pp 192-206.
